Вы находитесь на странице: 1из 40

chapter 6

Saving, Investment,
and the Current Account

I n a completely closed economy, one that is cut off from the rest of the
world, aggregate saving would necessarily equal aggregate investment.
The output in the economy is divided between current consumption uses
and investment, so that Q = C + I. At the same time, the income received
by households, which is also equal to Q, must he divided between con-
+
sumption and saving, so that Q = C S. We see immediately that I = S,
that investment must always equal saving. Both saving and investment
represent that part of national output which is not used for current con-
sumption.
Of course, the saving and the investment in a national economy are
not necessarily made by the same households (and firms).' Some house-
holds might want to save without having investment projects to under-
take, while other households might have investment projects but no sav-
ing. Financial markets solve the problem of getting the saving to those
who want to invest. Through them, the savers would accumulate financial
assets while the investors accumulate financial liabilities. To take one
simple example, the investors could issue bonds to finance their invest-
ments that could be purchased by the households that seek to save.
In an open economy, however, in which a nation's residents trade
goods and financial assets with residents in other economies, it is no
longer true that a nation's saving must always equal the investment that
takes place within the country. A nation's households might want to save
more than they want to invest at home, with the excess saving lent to
investors in other countries. In this case, the country would accumulate
net financial claims against residents abroad.
But what happens to the national output that is produced but neither
consumed nor invested? It is exported to foreigners. As we shall see,
there is an intimate relationship between the saving-investment balance of
a country and the net exports of the country.

'For analytical purposes, it is not necessary to distinguish at this point between


households and firms, so our discussion is in terms of households for simplicity.
150 Part I1 Intertemporal Economics

In this chapter, we study the determinants of national lending and


borrowing from the rest of the world. The current account of the balance of
payments is the key concept at the center of our discussion. (In addition to
the economic analysis of the current account given in the chapter, we also
discuss the accounting of the current account balance in an appendix.) When
residents in one country lend more to foreigners than they borrow, and thus
accumulate a net financial claim against the rest of the world, we say that the
country has a current account surplus. When the country is accumulating a
net liability (or running down its net claims) against the rest of the world, the
economy has a current account deficit. A current account surplus exists
when national saving exceeds national investment (with the difference lent
abroad), and a current account deficit exists when national investment ex-
ceeds national saving. We shall see that the current account balance is
closely related to the net export balance.
The current account has a crucial intertemporal dimension. The econ-
omy a s a whole, like the individual households (and firms) that compose the
economy, has an intertemporal budget constraint. If the economy runs a
current account deficit today, its residents are increasing their net debt to the
rest of the world. Eventually, the country will have to cut back on domestic
consumption in order to pay the interest on the accumulated debts. As
domestic consumption is cut back, national output that was used for con-
sumption is increasingly used for net exports. As we shall see, the country's
net exports are, in essence, its method of paying the interest burden on the
liabilities accumulated while running current account deficits.

6-1 A FORMALANALYSIS INVESTMENT,


OF SAVING,
ACCOUNT
AND THE CURRENT

We now turn to a formal model of the current account. In order to simplify


the theory, we continue (as in the past two chapters) to assume a classical,
fully employed economy, with a stable price level for goods (P = 1). (Later
on in the hook, we will discuss the implications for the current account of
aggregate demand-induced fluctuations in output within the framework of
the Keynesian model.)
In a closed economy, saving must equal investment. Since both saving
and investment are a function of the domestic interest rate, r, we can draw
the saving and investment schedules, as we do in Figure 6-1, with saving an
increasing function of r and investment a decreasing function of r.2 Of
course, saving and investment are also functions of many other things: cur-
rent and future income, expected profitability, and so on. These other fac-
tors are held fixed in the background when we draw saving and invest-
ment schedules like those in the graph. The domestic interest rate adjusts
to equilibrate saving and investment at the level given by the equilibrium
point E.
We can see the effects of various kinds of shocks on domestic saving,
investment, and interest rates quite clearly. Consider, for example, the ef-

Remember from Chapter 4 that the effect of a rise in the interest rate on saving is
ambiguous because the substitution effect tends to cause saving to rise while the income effect
may cause saving t o fail. As we said in Chapter 4, we take the normal case to be one in which a
rise in interest rates is associated with a rise in saving.
Chapter 6 Saving, Investment, and the Current Account 151

Figure 6-1
Saving, Investment, and the
Interest Rate in a Closed
Economy

fects of a temporary increase in output resulting from a favorable supply


shock-a bountiful harvest, for example. Households will want to save more
at any given interest rate, so that the saving schedule will shift to the right, as
shown in Figure 6-2a. The investment schedule will not shift, however, if the
output change is strictly temporary. As long as the future production func-
tion remains unchanged, the desired capital stock in the future also remains
unchanged. Thus, the I schedule will not shift. The result of the temporary
output increase, therefore, is afall in interest rates and an increase in current
saving and investment, as the equilibrium shifts from E to E' in Figure 6-2a.
Now let us consider the effects of an anticipated future increase in
income, one that also shifts upward the marginal productivity of capital in
the future. In this case, current saving will tend to fall, as households borrow
against their higher future income; investment will tend to increase as well,
to take advantage of the higher marginal productivity of capital. The result is
shown in Figure 6-2b, as a leftward shift in the saving schedule and a right-
ward shift in the investment schedule. We can say with certainty that inter-
est rates will rise, while overall saving and investment might rise or fall.

Figure 6-2
Effects of Economic Shocks o n Saving- and Investment in a Closed
Economy
152 Part I1 Intertemporal Economics

Most economies in the world are not closed, however, so the assurnp-
tion that a country's saving and investment must always balance is not very
useful. Residents in one country can generally lend or borrow from the rest
of the world, and thereby build up claims or liabilities vis-a-vis residents in
other countries. Thus, saving and investment analysis must be expanded to
take into account the international flows of financial assets.
Let B* be the net claims of a country's residents on residents in the rest
of the world. (We use the asterisk in general to denote a "foreign variable."
The asterisk here stresses that B* is a claim on foreign output.) B* is some-
times called the country's nef international inuestmenf position, or the net
foreign asset position. It may be thought of as an asset in the form of a bond
(hence, the notation), though in fact claims against the rest of the world can
be held in many forms: bonds, money, equities, and so on. B* measures the
assets of national residents vis-a-vis foreigners, minns the liabilities vis-a-vis
foreigners. When B* is positive, the country is a nef credifor of the rest of
the world, and when B* is negative the country is a net debtor of the rest of
the world.
We define the country's current account (CA) as the change in its net
financial asset position with respect to the rest of the world:
CA = B* - B!l (6.1)
Note that current account surpluses imply an accumulation of foreign assets
or a reduction in foreign liabilities. Deficits imply a decumulation of foreign
assets or an increase in foreign liabilities.
Equation (6.1) tells us that the current account in this period (CA) is the
change in net foreign assets, which we denote by B*, between this period
and the previous period (denoted by the subscript -,). Notice that the level
of B* in a given period is a result of past current account surpluses and
deficits. Starting from an initial year (arbitrarily denoted year O), the net
international investment position of a country in year f (BT) is equal to B,*
plus the sum of current accounts in the years between 0 and f :

In many countries, especially in the developing world, B* is a negative


number, because the countries' current accounts have been negative for a
long time. The situation of the heavily indebted developing countries has
generated much attention and discussion during the past decade and has
become known as the Third World debt crisis. We analyze this crisis in detail
in Chapter 22.
Table 6-1 presents the evolution of the U.S. current account and its net
foreign asset (NFA) position since 1970.3The NFA measures the creditor or
debtor status of the United States vis-a-vis the rest of the world; that is, it
measures the balance of total foreign assets minus total foreign liabilities.
Note that the current account deficits of the United States during the
1980s have transformed this country from the major international creditor
country to the world's biggest net debtor. Indeed, by the end of 1988 the

According to current usage, we will treat the terms "net international investment
position" and "net foreign asset position" as synonyms.
C h a p t e r 6 Saving, Investment, a n d t h e C u r r e n t A c c o u n t 153

Current Account Net International


Year Balance Investment Position

1970 2.3 58.6


1971 -1.4 56.1
1972 -5.8 37.1
1973 7.1 61.9
1974 2.0 58.8
1975 18.1 74.6
1976 4.2 82.6
1977 - 14.5 72.4
1978 - 15.4 76.7
1979 -1.0 95.0
1980 1.1 106.3
1981 6.9 140.9
1982 -5.9 136.7
1983 -40.1 89.0
1984 -99.0 3.3
1985 -122.3 -111.4
1986 -145.4 -267.8
1987 -162.3 -378.3
1988 -128.9 -532.5
1989 -110.0 -

Source: Economic Report of the President, 1991, (Washington,


D.C.: U.S.Government Printing Ofice, 19911, Tables 8-101
and B-102, various issues.

United States had accumulated over $500 billion of net foreign liabilitie~.~
This is more than three times the debt of Brazil or Mexico, the largest
developing country debtors. In spite of its size, however, the U S . problem

' W e should acknowledge at this point that the data underlying Table 6-1 have many
measurement problems. Some authors have argued that the United States did not really become
a net debtor during the 1980s because the value of U.S. assets abroad is much higher than the
official data indicates. For one thing, U.S. investments abroad have been measured traditionally
at historic cost. Other kinds of errors in the data do tend to understate U.S. debts to foreigners,
however. While we cannot be sure of the overall level of the net U.S. debt. it is surelv the case
that the U.S. net international investment position fell sharply in the 1980s, moving from a large
surplus to a much smaller surplus or to a deficit. Indeed, a recent estimate based on the market
value of investments revised the U.S. NIIP upward to $ 2 6 8 billion in 1989, still a sizable
number, but much less than the $ 5 3 2 billion shown in the table.
154 Part I1 Intertemporal Economics

is of a smaller magnitude relative to income; the net international liabilities


of the United States represent only about 10 percent of its GDP, while
Mexico's net debt is well over 50 percent of its GDP.
Note also that the current account is not exactly equal to the change in
net foreign assets. A variety of factors account for this discrepancy: unre-
corded capital which sometimes show up in the balance of payments
under the category "errors and omissions," valuation changes on existing
assets and liabilities which affect the net asset position but not the current
account, expropriations of foreign assets, and defaults on international debt.
To show how the current account is related to saving and investment,
we must first look back at the budget constraint of an individual household.
Recall from equation (5.5) of the previous chapter that for a given household
i, the change in financial assets is equal to the difference between the saving
and investment of the household:

Now, write the household's income as Yi = Q + r B L l , and use the fact that
saving Si equals Yi - C' to find

An individual household can hold claims against other domestic households


or against foreigners. If we add up all the net claims of households to get the
net asset position of the entire economy, the claims that are owed by one
household to another net out of the sum of all households, since claims
between households are assets for some households but equal liabilities for
others. What remains are the net claims of the economy against the rest of
the world, which we have denoted B*. Thus, in adding (6.4) over all house-
holds we find for the economy as a whole

Once again substituting Y = Q + rB" (GNP = GDP + net income from


abroad), and S = Y - C, we can now write

Equation (6.6) can be interpreted quite simply. Since it can be rewritten as


S = I + (B* - B"), it tells us that domestic saving can be used for two
things: domestic investment ( I )or net foreign investment (B* - BZ1).
Equations (6.1) and (6.6) make it clear that the current account can be
expressed as the difference between national saving and investment:

As long as domestic residents can borrow and lend from foreign residents,
domestic saving and investment do not have to be equal. The difference
between saving and investment is precisely measured by the current account
balance. Clearly, in a closed economy, the current account concept is irrele-

In industrialized countries, these unrecorded capital flows are known by the elegant
technical term portfolio reallocations. In the developing world, these flows are called capitul
Jight, a term that has a distinctly negative connotation. We analyze the issue of capital flight in
Chapter 22.
Chapter 6 Saving, Investment, and the Current Account 155

SAVING,
INVESTMENT,A N D THE CURRENT ACCOUNTI N THE UNITED STATES,
1950- 1990
(AS P E R C E N T A G E O F GDP)

Gross
Private Balance
Gross Domestic Saving - on Current Statistical
Year Saving Investment Investment Account Discrepancy

p = preliminmy.
D.C.:U.S. Government Printing*OfJice, IYYI), Tuh1e.r
Source: Economic Report of the President, 1991 (Wu.c.hin~ton,
B-28und B-102.

vant. In an economy completely isolated financially from the rest of the


world (no net claims), the current account is always zero.
Table 6-2 presents the time series of saving, domestic investment, and
the current account as a percentage of GDP for the United States in the
period 1950-1990.Wotice that in the 1950s, 1960s, and 1970s the United
States consistently experienced a current account surplus. This situation
was sharply reversed in the 1980s. During 1981-1990, the average current
account deficit was about 2 percent of GDP. Interestingly, the current ac-
count decline in the 1980s was due to a sharp fall in the national saving rate
rather than to a rise in domestic investment. In fact, domestic investment

We should point out one data distortion in Table 6-2. In the United States, gross saving
is measured as the sum of private saving and the public surplus. The public surplus is govern-
ment saving minus government investment. Thus, government investment is subtracted from
gross saving rather than being included in total investment, which includes only private capital
formation. Thus, for this reason, the data understate both saving and investment by misclassify-
ing government investment. Nonetheless, even if government investment were properly classi-
fied, the direction of change in recent years (toward lower national saving and a lower current
account balance) would remain.
156 Part I1 Intertemporal Economics

The six major OECD trading partners of the U.S. are Canada, France, Germany, Italy, Japan, and the United Kingdom.
The figure shows the sum of their current accounts, measured in dollars, as a percentage of their combined GDP,also
measured in dollars.

Figure 6-3
The Current Account in the United States vis-a-vis Other Industrialized
Countries
(From International Monetary Fund, International Financial Statistics.)

also declined in this period, but the drop in saving was even more pro-
nounced. (In turn, most of the fall in national saving is due to the behavior of
the public sector, a point we study in greater detail in the next chapter, when
we analyze formally the role of the government sector.)
As the current account balance of the United States declined during the
1980s, the rest of the world had to be running a current account surplus vis-a-
vis the United States. After all, the world as a whole is a closed economy.
Figure 6-3 depicts the behavior of the current account in the United States
vis-a-vis the rest of the member countries in the Organization for Economic
Cooperation and Development (OECD).7 The graph clearly shows the nega-
tive relationship between the U.S. current account and that of the rest of the
OECD countries. As the U.S. CA deficit increased to over 3 percent of
GDP, the CA surplus among the other OECD members reached over 2
percent of their combined GDP.
Of course, the average for the other 23 OECD member countries hides
important differences in individual behavior. Indeed, this average is strongly
influenced by the vast current account surpluses in Japan and West Germany
as well as the relative weight of their economies within the OECD. Table 6-3
shows a breakdown of the current account behavior for the United States

' The OECD is an association of 24 major industrialized countries. These are Australia,
Austria, Belgium, Canada, Denmark, Finland, France, Germany, Greece, Iceland, Ireland,
Italy, Japan, Luxembourg, the Netherlands, New Zealand, Norway, Portugal, Spain, Sweden,
Switzerland, Turkey, the United Kingdom, and the United States.
Chapter 6 Saving, Investment, and the Current Account 157

United United
Year States Canada France Germany Italy Japan Kingdom

Billions of U.S. Dollars


-4.2 - 14.0
-4.8 -3.4
-12.0 5.0
-5.2 5.4
-0.9 9.7
0.0 17.0
2.4 40. I
-4.4 46.1
-3.5 50.5
-4.3 55.7
Percentage of GDP
-0.6 -1.7
-0.8 -0.5
-2.2 0.8
-1.0 0.8
-0.2 1.6
0.0 2.7
0.3 4.5
-0.5 4.1
-0.4 4.2
-0.4 4.7

Source: International Monetary Fund, International Financial Statistics, selected issues.

and its six major OECD trading partners, Canada, France, Germany, Italy,
Japan, and the United Kingdom, for the 1980s.
There is yet another way to express the current account. Notice from
equations (6.3) and (6.7) that
CA= Y - ( C + Z ) (6.8)
We define "absorption," A, as the sum of consumption and investment, that
is, total spending by the domestic resident^,^ or
A=C+Z (6.9)

Strictly speaking, consumption and investment should be interpreted here to include


government consumption and investment. We explicitly introduce the government in our frame-
work in the next chapter.
158 Part I1 Intertemporal Economics

Therefore, the current account is also the difference between income


and absorption:

This was an important insight of Sidney Alexander of the Massachusetts


Institute of Technology in the early 1 9 5 0 ~ . ~
Equation (6.10) has a very intuitive appeal. Countries run current ac-
count deficits when they spend (or absorb) more than they earn. This re-
quires them to run down their foreign assets or to increase their net liabilities
to the rest of the world.
Thus, a current account deficit occurs when a country "spends beyond
its means" (absorption is greater than income) or when it "invests in excess
of its own saving." While these two ways of measuring the current account
are equivalent, they certainly conjure up different value judgments about a
current account deficit. When economists want to complain about a current
account deficit, they tend to say that the country is living beyond its means;
when they want to defend a current account deficit, on the other hand, they
say that the country's investment climate is highly favorable (causing invest-
ment in excess of national saving). Of course, a current account deficit is-
by itself-neither a good nor a bad thing. The appropriateness of the current
account position must be evaluated in terms of the intertemporal prospects
facing an economy.
Using the diagram in Figure 6-1, we can readily see how the current
account is determined. There we depicted saving as an increasing function of
the interest rate and investment as a decreasing function of the interest rate.
In the closed economy, the interest rate adjusts to equilibrate saving and
investment, as we saw earlier.
Now, let us assume that the economy is open and that in fact its
residents can borrow and lend freely at a given world interest rate, which we
call r. In effect, we are making the small-country assumption, that saving
and investment decisions in the home country-whose saving and invest-
ment are depicted in Figure 6-4a-do not affect the world interest rate. For a
given world r, saving and investment need not be equal in the country, with
the gap reflecting the current account deficit or surplus of the country. If the
world interest rate is relatively high, say, rH in Figure 6-4a, saving in the
domestic economy will be higher than investment and the country's current
account will be in surplus. (The current account surplus is measured by the
horizontal difference between the S and I schedules at the rate rH.) Con-
versely, if the world interest rate is relatively low, say, r,, investment will
exceed saving in the domestic economy and the economy will be running a
current account deficit.
Using this simple framework, we can represent the current account as
an increasing function of the interest rate, as shown in Figure 6-4b. At every
interest rate, the horizontal difference between the saving and investment
schedules in Figure 6-4a measures the current account. By shifting the inter-
est rate, we can draw the curve CA in Figure 6-4b. Notice that the CA
schedule is always flatter than the S schedule, because a higher interest rate

Sidney Alexander, "The Effects of Devaluation on a Trade Balance," International


Monetary Fund Staff Papers, 1952, pp. 263-278.
Chapter 6 Saving, Investment, and the Current Account

CA

(a)

Figure 6-4
Saving, Investment, and the Current Account

not only increases saving but also reduces investment, and both of these
effects improve the current account. In a later section, we examine more
carefully those factors that are likely to affect a country's current account
balance.

6-2 THECURRENT
ACCOUNT
AND INTERNATIONAL
TRADE
So far, we have described the current account without mentioning interna-
tional trade. This may be surprising, since most of us typically think about
the current account as a trade phenomenon, a matter of exports and imports.
In truth, there is an intimate link between the saving-investment balance and
the export-import balance, a link which leads us to a more subtle under-
standing of current account imbalances.
When a country absorbs more than it produces (A > Q ) , it is using
more resources than are available to it from domestic production alone. A
country can do that only by importing goods from the rest of the world. More
precisely, the country must import more from the rest of the world than it
exports to the rest of the world, so that on balance it is receiving real
resources from abroad. For this reason, a current account deficit tends to be
associated with an excess of imports over exports, and a current account
surplus tends to be associated with an excess of exports over imports. Now
let us look more closely at this relationship.
For a given amount of total domestic absorption A , the total spending
is divided between absorption on domestic goods (denoted Ad) and absorp-
tion on imports (denoted ZM):I0

lo Technically, all goods will be measured in units of the home goods. That is, IM
signifies the nominal value of total imports divided by the price index of the domestic output.
160 Part 11 Intertemporal Economics

At the same time, all goods produced at home must be either sold
domestically (in the amount Ad) or exported, X. Thus,
Q=Ad+X (6.12)
The country's trade balance is measured as the value of exports minus
the value of imports (TB = X - IM). But since exports are equal to total
output minus the portion of it that is consumed domestically ( X = Q - Ad),
we can conclude that
TB=X-IM=Q-Ad-IM=Q-A (6.13)
Now, with the trade balance equal to output minus absorption, and
with the current account equal to income minus absorption, the difference
between the trade balance and the current account balance is the net factor
payments from abroad (NF). In our model, N F is simply the interest pay-
ments on net foreign assets, equal to rB!, ." Consequently, because CA =
Y - A , we also can write CA = Q + rBf - A. Then, using (6.13), we have
CA = TB + rBTI (6.14)
Under ordinary circumstances, rBfl is small relative to the trade bal-
ance, in which case the current account balance and the trade balance are
nearly the same. Current account deficits often signal not just an excess of
investment over saving, or absorption over income, but also an excess of
imports over exports. It is possible, however, to have a current account
deficit with a trade-balance surplus (or vice versa), if the earnings on the net
foreign assets are relatively large. In Box 6-1, we examine the current ac-
count balances of several countries to see how the overall balances actually
depend on trade, interest payments, and other items.
To summarize, there are four different ways to describe the current
account: (1) as the change in net foreign assets (CA = B* - Bf,), (2) as
national saving net of investment (CA = S - I ) , (3) as income minus absorp-
tion (CA = Y - A), and (4) as the trade balance plus net factor payments
from abroad (CA = X - I M + NF).
In past years, some economists have argued as if these different defini-
tions hinted at different "theories" of the current account, including an
intertemporal theory that stresses saving and investment; an "elasticity ap-
proach" that stresses factors determining imports and exports; an "absorp-
tion approach" that stresses the determinants of absorption relative to in-
come; and so forth. This debate among the various schools of thought has
been fruitless, however. All formulations of the current account are equally
true, and all are linked together by simple accounting identities. There is no

In real life, though not in our simplified model, there are some other items that cause
the current account and the trade balance to differ. For example, the receipt of foreign aid from
abroad raises the current account relative to the trade balance, though the foreign aid is a form
of transfer payment and not a receipt of income on foreign assets. A complete breakdown of the
difference in the two balances is shown in Table 6-4. Note that the difference between the trade
balance and the current account balance includes two categories: "other goods, services, and
income" and "unilateral transfers." The "other goods, services, and income" category in-
cludes income received on net foreign assets (the subcategory "interest and dividends"), as
well as receipts on travel (tourism), workers' remittances, and some other items. The "unilat-
eral transfer" category includes foreign aid as well as transfers from the private sector.
Chapter 6 Saving, Investment, and the Current Account 161

Box 6-1
What Is Hidden by a Summary
Current Account Statistic?
Cross-country comparisons of current account behavior, like those
shown in Table 6-3, often do not go beyond the most aggregated level: the
ratio of the current account to GDP. But while this summary statistic
conveys some important information, it also hides quite a bit. Is a given
current account deficit due to high investment levels or to low saving? Is
the explanation for the deficit found in a trade deficit or in high interest
payments on foreign debt? The answers vary substantially across coun-
tries, a point that shows up clearly in Table 6-4. (For a discussion of
balance-of-payments accounting, see also the appendix to this chapter.)
Consider the current account balances of various countries in 1989.
Notice that in the United States the current account deficit is almost
completely accounted for by the trade deficit, with the rest of the current
account close to equilibrium. In Japan, a major trade surplus is the princi-
pal cause of the current account surplus. Behind these figures, the coun-
try runs an important deficit on services, as the Japanese have grown fond
of traveling abroad. Of course, this situation is not static. The United
States used to have a huge surplus in services due to high profit remit-
tances and interest payments on U.S. loans abroad. Persistent current
account deficits have deteriorated its net foreign asset position, however,
as we saw in Table 6-1, and this has obviously reduced the net income
from capital. Japan has persistently accumulated net foreign assets during
the period, and thus it presents the opposite case.
For major debtors, such as Brazil in the table, the current account
shows a relatively small surplus of $4.1 billion despite a massive trade
surplus of $19.1 billion. Huge interest payments on foreign debt account
for most of the discrepancy. Because of this factor, in 1988 Brazil had a
current account deficit together with a huge trade surplus. Among service
payments, workers' remittances are a very important source of foreign
exchange on the current account for countries such as Turkey, and, to a
lesser extent, the Philippines. Notice also that workers' remittances rep-
resent a substantial outflow of funds from the United States and Japan, as
foreign workers attracted by high wages in the United States send money
back to their families in their native countries. In other countries such as
Spain and Thailand, tourism is a major foreign-exchange earner. Indeed,
tourism is the most important source of foreign exchange in Spain, with a
net contribution of over $13 billion in 1989, or about 30 percent of the
country's exports.
A final group of countries get much of their foreign exchange on
current account through unilateral transfers, that is, gifts from other na-
tions. China, India, Indonesia, Bangladesh, Egypt, and Israel collect the
most dollars through official development assistance, as shown in Table
6-5. However, if the ranking is made as a proportion of the recipient
country's GDP, the top 1 1 earners of grants are in Africa.
THECOMPOSITION
OF THE ACCOUNTBALANCEFOR SELECTED
CURRENT DECEMBER
COUNTRIES, 1989
L
(MILLIONS OF U.S. DOLLARS) m
0
0

United Philip- Thai- o


4
States Japan Brazil* Mexico Turkey pines Spain land E.
0
V)

Trade balance
Exports
Imports
Other goods, services,
and income
Travel
Interest and dividends
Workers' remittances
Other
Unilateral transfers
Official transfers
Other
Current account balance

* December 1988.
Source: International Monetary Fund, Yearbook, Balance of Payments Starisrics, 1990.
Chapter 6 Saving, Investment, and the Current Account 163

Amount
(millions of US$) As a % of GDP

China Mozambique
India Somalia
Indonesia Tanzania
Bangladesh Lesotho
Egypt Malawi
Israel Chad
Pakistan Mali
Kenya Lao PDR
Tanzania Mauritania
Philippines Burundi
Sudan Central African
Mozambique Republic
Nepal

Source: World Bank, World Development Report 1991 (New York: Oxford University Press,
1991), Table 20.

separate "intertemporal theory" or "trade theory" of the current account.


Each of the approaches, properly specified, must lead one back to the same
more basic considerations.

In this section we study at greater length the factors that influence the
current account balance in a small country facing a given world interest rate.
We focus on the effects of different shocks that may hit the economy-
changes in world interest rates, fluctuations in the terms of trade, and invest-
ment movements.

World Interest Rates


The first factor of importance is the world interest rate itself. Note in Figure
6-4 that as the world interest rate rises from r, to r h , domestic investment
falls, saving rises, and the current account moves to a surplus. Thus, there is
a positive relationship between the current account for the small open econ-
omy and the world interest rate at which its residents borrow and lend.
Remember that current account changes have effects on both financial
and trade flows. Suppose for a moment that the economy starts from current
account balance at point r, in Figure 6-4. A rise in interest rates makes the
current account go to a surplus, as consumers save more (consume less) and
invest less out of the fixed amount of national income. The decline in domes-
164 Part I1 Intertemporal Economics

tic absorption means that imports fall and that a greater amount of domestic
production is available for export. Thus, the shift to current account surplus
also implies an increase in net exports, a trade phenomenon. The Jinancial
counterpart of the improvement in the trade balance is the accumulation of
net foreign assets, B*.

Investment Shocks
Suppose that investment prospects improve in a small economy that faces a
given world interest rate. In Figure 6-5, this is represented as a shift to the
right in the investment schedule. If the economy started from equilibrium at
point A , the current account moves to a deficit of magnitude AB. In Figure
6-2, the effect of the investment shock on a closed economy, was mainly to
raise interest rates. Here, in an open economy, the domestic interest rate is
given by the world rate. Thus, an investment surge has a deteriorating effect
on the current account, while interest rates remain unchangeil.
A good example of such a phenomenon took place in Norway after the
major world oil price increase in 1973. The oil shock made it highly profitable

Figure 6-5
The Current Account and
Improved Investment
Opportunities

to invest in oil exploration and development in the North Sea. Norway's


investment-to-GDP ratio, which had averaged 28 percent during 1965-1973,
increased by a full 10 points to 38 percent during 1974-1978. Most of this
surge in capital formation went to energy and energy-related ventures, in-
cluding the oil and gas pipeline between Norway and West Germany. Be-
cause the country's saving rate changed little (and even fell a bit), however,
the result of this investment surge was a massive current account deficit,
which reached almost 15 percent of GDP in 1977.12

Output Shocks
In many countries, output occasionally drops temporarily because of unfa-
vorable weather conditions or other exogenous shocks to a major sector of
the economy. Take the case of an agricultural country hit by a severe

l 2 For an analysis of current account behavior in both industrialized and developing


economies, see Jeffrey Sachs, "The Current Account and Macroeconomic Adjustment in the
1970s," Brookings Papers on Economic Activiry, 1: 1981.
Chapter 6 Saving, Investment, and the Current Account 165

I, The Current Account and a


s, I Transitory Output Decline

drought, or a Caribbean country hit by a hurricane. The life-cycle theory of


saving predicts that people want to maintain a stable consumption level
despite the temporary decline in output, and thus aggregate saving will de-
cline in response to the shock. For a given amount of investment, the current
account will deteriorate, as Figure 6-6 shows. If the country started from
equilibrium at point A , the current account deficit after the temporary shock
is AC in the graph. (Remember from Figure 6-2 that in response to a tempo-
rary adverse shock, the closed-economy response is a rise in interest rates,
and some decline in domestic investment.)
If the shock is permanent, however, then saving should not fall signifi-
cantly in response to the shock. Instead, it makes more sense to reduce
consumption by the amount of the fall of output when the decline in output is
permanent. Thus, with a permanent decline in output, the current account
does not shift into deficit. (In fact, if investment demand falls in response to
some long-term adversity, the current account might actually turn to surplus
despite the decline in current output.)

Terms-of-Trade Shocks
The terms of trade, which will be denoted as TT, is the price of a country's
exports relative to the price of its imports (TT = PxlPM). Because countries
export more than a single good, Px should be interpreted as a price index for
all export goods. The same applies to PM.A crucial aspect of terms-of-trade
changes is that they cause income effects for the country, effects that are
akin to shifts in national output. A rise in the terms of trade means that Px
has gone up relative to PM.With the same physical quantity of exports, the
country can now import more goods. The country's real income rises be-
cause of the greater availability of imports.I3
A transitory rise in the terms of trade implies a transitory increase of
income relative to permanent income. Consequently, aggregate saving in the
country will tend to rise because of consumption-smoothing behavior. Start-

l 3 A simple measure of the percentage rise in real income caused by the change in TT is
found as follows: multiply the percentage change in the terms of trade by the share of imports in
GNP. Thus, if the terms of trade improves by 10 percent, while the import-GNP ratio is 20
percent, the terms of trade improvement is akin to a 2 percent (20 percent times 0.10) improve-
ment in real national income.
166 Part I1 Intertemporal Economics

ing from equilibrium, the current account will tend to move to a surplus.
Following a permanent rise in the terms of trade, however, households will
adjust their real consumption upward by the amount of the terms-of-trade
improvement. Saving rates do not necessarily rise in this case, and the
current account does not necessarily move toward or into surplus.
Colombia, for example, has experienced sizable temporary income
fluctuations when the price of its major export, coffee, has changed relative
to other prices. In the late 1970s, a major rise in the relative price of coffee
had significant effects throughout the economy. The macroeconomic result
was true to the theory. Domestic saving went up as a proportion of GDP and
the current account improved ~ignificantly.'~
The theory of the current account, therefore, offers an important pre-
scription for the "optimal" response to fluctuations in the terms of trade. If a
change in the terms of trade is temporary, it should be absorbed by changes
in the current account; that is, a terms-of-trade improvement should result in
a surplus, while a terms-of-trade decline should lead to a deficit on the
current account. If a change in the terms of trade is permanent, households
adjust their consumption levels in response to the shocks so that saving rates
do not fluctuate. Permanent shifts in the terms of trade should therefore have
little effect on the current account (except as the TT shock might affect
investment spending).
This basic wisdom is sometimes encapsulated in the phrase "finance a
temporary shock; adjust to a permanent shock." The term "finance" here
means to borrow or lend-to run current account surpluses or deficits-in
response to transitory disturbances; the term "adjust" means to vary the
consumption level up or down in response to permanent TT shocks. This
general principle is a fundamental guidepost for the lending policies of the
International Monetary Fund (IMF). The IMF was formed immediately after
World War I1 to assist countries with external payments difficulties and to
promote international stability in the monetary system. In 1962, the IMF
created the Compensatory Financing Facility (CFF), a loan fund designed
explicitly to make loans to countries suffering temporary shortfalls in export
earnings. To qualify for a CFF loan, the country must demonstrate in precise
detail both that it has suffered a decline in export earnings and that the
shortfall is temporary. If the shortfall appears to be permanent, then the IMF
does not make a CFF loan, and instead advises the country to cut back on
spending levels to match the shortfall in its exports.
The idea of financing a temporary shock but adjusting to a permanent
shock represents both a "normative" theory (what should happen) and a
"positive" theory (what will happen) of the current account. But, as we shall
see, positive theory sometimes falls short of predicting what actually hap-
pens to the current account. The positive theory of the current account
depends on various assumptions: that economic agents are rational, inter-
temporal optimizers; that they are able to distinguish temporary from perma-
nent shocks; and that they are able to borrow and lend freeiy in response to
those shocks. We shall soon see that these assumptions may well be violated

l4 Sebastian Edwards has studied this interesting experience in his article "Commodity
Export Prices and the Real Exchange Rate in Developing Countries: Coffee in Colombia," in S .
Edwards and L. Aharned, eds., Economic Adjustment and Exchange Rates in Developing
Countries (Chicago: University of Chicago Press, 1986).
Chapter 6 Saving, Investment, and the Current Account 167

in real economies. In particular, when governments act as borrowers and


lenders, they often fail to act as farsighted intertemporal maximizers.
Thus, when many developing countries enjoyed large terms-of-trade
improvements at the end of the 1970s, they failed to run current account
surpluses as theory predicted they would. Instead, the governments in these
countries often acted as if the terms-of-trade improvements were permanent
instead of transitory, and they raised spending by the full amount of the real
income gain, even though the gain was likely to be short-lived. Mexico, for
example, spent the huge windfall in oil export earnings that arose when oil
prices shot up during 1979 and 1980. Once the terms of trade reversed in the
early 1980s, Mexico and other such governments found themselves with
unsustainably high spending levels and big political difficulties in cutting
spending back down to manageable levels. Often it took a deep crisis-
economic and political-before government spending levels were cut back
to sustainable levels. (We will discuss some of these adjustment problems in
the next chapter, and again in Chapter 22, when we analyze the developing
country debt crisis.)

We have seen that personal saving and investment decisions in a particular


period influence someone's future path of consumption and income. A per-
son who borrows today must consume less than his or her income in the
future in order to repay the loan. Similarly, the levels of national saving,
investment, and the current account influence the future path of consump-
tion and income for the economy as a whole.
Suppose that a natural disaster makes output fall temporarily in the
current year. A decline in the country's output translates into lower income
for the average household. As individual households attempt to smooth their
consumption by borrowing against higher future income, aggregate saving
declines, and the national economy experiences a deterioration in the cur-
rent account. The country then borrows from abroad, or at least runs down
its existing stock of foreign assets. In the future, it will have to consume less
than income in order to repay the debts incurred today.
An example of this phenomenon was Ecuador in 1987. When a major
earthquake destroyed 35 kilometers of the country's oil pipeline, it left oil
production interrupted for five months. Oil is Ecuador's principal export,
and the earthquake produced a sharp but temporary decline in the country's
income. Consequently, national saving collapsed and the current account
reached a deficit of about 12 percent of GDP. Following that crisis and the
foreign borrowing that it provoked, Ecuador will have to restrict consump-
tion to service the debts incurred during that year.

The Intertemporal Budget Constraint in the Two-Period Model


We can also examine the country's intertemporal budget constraint formally
using the two-period model. Suppose, as we did at the household level, that
the country starts with no foreign assets (B,* = 0). In that case, the value of
B* in period I (B? is equal to the current account surplus in the first period:
168 Part I1 Intertemporal Economics

The change in net foreign assets from the first to the second period is
the current account balance in the second period:

But under the rules of a two-period model, the country must end with
no net foreign assets (B,* = O), and it undertakes no investment in the second
period (I2 = 0). Therefore, we can combine equations (6.15) and (6.16) to
obtain

Thus, we see that what was true for individual households is also true
for the nation as a whole. Countries too are bound by a national intertem-
poral budget constraint: the discounted value of aggregate consumption
must be equal to the discounted value of national production net of in-
vestment.
Take a simple case where there are no attractive investment opportuni-
ties. Under these conditions, the economy's only decision is how much to
consume today and how much to save. In Figure 6-7, the country's budget
constraint is shown by the line CC. For all the points on CC, CI +
C21(1 + r) = QI + Q2/(l + r). To the southeast of point Q, the economy
would be running a current account deficit in the first period, with Cl > Ql .
To the northwest of point Q, the country would be running a current account
surplus. The point where the economy will actually locate along the budget
line depends on the preferences of the society.
Three fundamental conclusions emerge from this analysis:
1. If consumption is greater than output in the first period (CI > e l ) ,
then consumption has to be smaller than output in the second period
(C2 < Q2). The reverse is also true: if C1 < Q l , then C2 > Q2.
2. Since, in the absence of investment, the trade-balance surplus is the
difference between output and consumption (TBI = Ql - CI), then
the trade deficit in the first period must be matched by a trade
surplus in the second period.

Period 2
C2, Qz I

CA surplus

Q2 -------- Figure 6-7


The Country's Budget
C I ,Q I
Constraint and the Current
c Period 1 Account
Chapter 6 Saving, Investment, and the Current Account 169

3. If the country runs a current account deficit in the first period,


thereby incurring foreign debt, it must run a surplus in the future, in
order to repay the debt. Similarly, if it runs a surplus in period 1, it
must run a deficit in period 2.
Algebraically, we can state the country's intertemporal budget con-
straint in several analogous ways. First, we have seen that the discounted
value of consumption must be equal to the discounted value of output net of
investment. Second, we can rearrange terms in equation (6.17) to describe it
in terms of the trade balance in the two periods. Since TBI = Ql - CI - 11
and TB2 = Q2 - C2, it is easy to verify that the discounted value of the trade
balances has to be equal to zero:

This means that a trade deficit in the first period must be balanced by a trade
surplus in the second period of equal present value.
The third way of expressing the country's intertemporal budget con-
straint is in terms of the current account. Since an economy's current ac-
count is equal to the economy's accumulation of net foreign assets, we have
CAI = BY - B$ and CA2 = B: - BT. Assuming that the country starts with
no net foreign assets (BT = 0) and ends with no assets (B: = 0),we must
have
CAI + CA2 = 0 (6.19)
Before moving on, we should stress one key qualification. This analy-
sis assumes that a debtor always honors its debts, and the budget constraint
is derived under that assumption. There are cases, of great importance, in
which a debtor either cannot or chooses not to repay debts incurred in an
earlier period. In a domestic economy, debtors sometimes go bankrupt and
are unable to repay. In the international economy, where enforcement of
contracts is more difficult, debtors sometimes choose not to repay. In these
cases, the budget constraint may not be as stringent as most arguments
suggest. (We return to this issue at the end of the chapter, and again in
Chapter 22, when we discuss the developing country debt crisis.)
Let us consider a specific illustration of the intertemporal budget con-
straint to clarify concepts further. Suppose that the saving and consumption
preferences of individual households lead to a particular choice of consump-
tion in the CC schedule so that, say, C1 < Ql for the economy as a whole.
This situation is represented in Figure 6-8, and the appropriate balance-of-
payments accounts are shown in Table 6-6. The horizontal distance between
Ql and C1 measures the current account surplus and the trade-balance sur-
plus in period 1. Notice that there is no difference between the two measures
in this case. Why? Because the country starts with no net foreign assets.
Domestic households will be lending, on aggregate, an amount BT =
Ql - Cl to the rest of the world. This capital outflow exactly balances the
current account surplus. In the second period, the country consumes C2 >
Q2. The current account is in deficit while there is a capital inflow.
It is worthwhile to mention how these transactions would be recorded
in the balance-of-payments accounts kept by the government. (A detailed
description of balance-of-payments accounting is given in the appendix to
170 Part I1 Intertemporal Economics

Period 2 ,

Figure 6-8
The Budget Constraint and a
Contemporary Current
Account Surplus

this chapter.) The accounting of the balance of payments for this hypotheti-
cal country will look like the schedule shown in Table 6-6. We need to
introduce here just one new point in order to proceed. A capital outflow is
termed, by accounting convention, to be a deficit in capital account of the
balance of payments. (And a capital inflow is termed, by accounting conven-
tion, to be a surplus in the capital account of the balance of payments.) This
means that the current account and the capital account automatically add to
zero, as shown in the table.

The Intertemporal Budget Constraint with Many Periods


So far we have derived the intertemporal budget constraint in a two-period
framework, but it is easy to extend the analysis to many periods. For T
periods, with T > 2, we simply derive expressions that are analogous to
equations (6.17), (6.18), and (6.19), showing that the discounted value of
consumption must equal the discounted value of output net of investment,
that the present discounted value of trade balances must equal zero, and that
the current account balances between t = 0 and t = T must sum to zero.
The extensions from a two-period model to a T-period model are rather
straightforward. A new subtlety is added, however, in the (quite realistic!)
case that there is no known final period T at which all loans must be paid off.
If time just goes on without a final date, does this mean that a country can

Period 1 Period 2

Current account QI - -(QI - el>


Trade balance Ql - CI -(I + ~ N Q I- CI)
Service account 0 ~ ( Q -I el>
Capital account -(QI - (QI - CI)
Total (of current account
and capital account) 0
Chapter 6 Saving, Investment, and the Current Account 171

borrow any amount from the rest of the world, without concern for repay-
ments, knowing that it can always simply borrow more in the future to repay
any past debt? The answer is no. The international capital markets will still
require that the country live within its means, in that no lender will lend so
much to a country for which the only way to repay is to borrow the amount
due each period.
A scheme in which a borrower takes on too much debt (to increase
current consumption, for example), and then plans to repay it by borrowing
the money needed for debt servicing, is known as a Ponzi scheme.l5 Con-
sider what happens in such a scheme. Suppose the borrower owes a debt D.
When the debt D comes due, the borrower owes (1 + r)D. If it takes a new
loan equal to (1 + r)D to pay off the old lender, it now owes a larger amount
to the new lender. In the next period, the borrower will have to pay (1 +
r)2D, and again, it plans to borrow this larger amount to make the repayment.
In the following period, the borrower will owe (1 + r)3D. In each period,
then, the debt will grow at the geometric rate (1 + r).
Credit markets prevent this behavior (or they do not support such
behavior indefinitely): lenders require that a borrower's debt stay within
bounds, and at least they do not allow it to grow at the geometric rate (1 + r).
It can be proven mathematically that when the debt is constrained-by the
prudent behavior of lenders-to grow less rapidly than the geometric rate
(1 + r), the borrower is forced to live within its means in the sense that the
present discounted value of all its future consumption must equal the initial
wealth plus the present discounted value of all future output net of invest-
ment:

Let us define the country's net debt as D*, which is just equal to -B*.
In other words, when B* is negative (so that the country is a net debtor), D*
is positive. Now we can derive a very interesting equation. By bringing the
terms in (Q - Z) to the other side of the equation, and remembering that the
trade balance is equal to output minus absorption (TB = Q - C - I),we can
write (6.20) in the form

This very important relationship says that if a country is a net debtor,


and owes (1 + r)D,* in the first period, then the economy must run trade
surpluses in the future whose present discounted value (over the entire
future) equals the initial net debt. The country services its debt into the
future by a stream of trade-balance surpluses whose present value equals the
net debt that is owed to the rest of the world.
Be careful to interpret the condition established in equation (6.21) cor-
rectly, however. It does not require that a debtor country have a trade
surplus in every period, but only that the present value of all future trade
balances must be in surplus, equal to the value of the net debt. For example,

l5 After Charles Ponzi, a Boston wheeler-dealer, who became rich with a scheme of chain
letters in the 1920s.
172 Part I1 Intertemporal Economics

the United States at the end of 1988 had net foreign liabilities in the order of
$532 billion. This means that from 1989 onward, the United States will have
to run trade surpluses in present-value terms of $532 billion. Of course, this
does not imply that the United States will have to run trade surpluses in
every period.
Notice, however, another subtle point. Even though a country cannot
run a debt that grows forever at the rate of interest, it never has to pay off its
debt fully either. What is required is that the country pay interest on its
foreign debt (by running trade surpluses), not that the debt go to zero by
some specific date. Thus, a country could maintain a given net debt D each
year, and pay the interest due, rD, by running a trade surplus, without the
principal D ever returning to zero.
The intertemporal budget constraint of the country is sometimes stated
in terms of the net resource transfer (NRT) that a country must make. The
NRT measures the cash flow between the country and its creditors, and it is
measured as the net loans made to the country by its creditors, minus the
interest that the country pays on its foreign liabilities (what creditors
"take"). Thus the NRT in period t is given as
N R T = (D* - DT,) - rDT, (6.22)
Notice that in a "Ponzi scheme," the net resource transfer is precisely zero,
since the amount of new borrowing is just enough to pay the old debt: D* =
( I + r)D'F,, SO that NRT = 0.
Because the increase in net foreign debt (D* - D'F,) corresponds to the
current account deficit ( - C A ) , while the interest payments correspond to
the deficit on the service account, equation (6.22) can readily be restated in
terms of the trade balance:16
NRT = -TB (6.23)
Thus, when a country is running a trade-balance deficit, it is receiving a net
resource transfer from its creditors, and when it is running a trade-balance
surplus, it is making a net resource transfer to its creditors (in which case,
we sometimes say that the NRT to the country is negative).
Now the budget constraint for a debtor can be stated as the condition
that the negative of the discounted value of future net resource transfers
must equal the size of debt:

Obviously, this condition is the same as (6.21), since the NRT is equal to the
trade-balance surplus. Note also that the NRT condition rules out a Ponzi
scheme because in a Ponzi scheme, the NRT is always zero.
We should mention, once again, a limitation we pointed to earlier. The
"no-Ponzi scheme" condition is a plausible condition for capital markets,
but lenders do not always impose it successfully. Sometimes borrowers
are-inadvertently-allowed to borrow so much that they simply cannot

'"ctually, the NRT is equal to the trade balance plus nonfactor services (mostly tour-
ism, freight, and insurance). For simplicity, we have not considered in our analysis nonfactor
services. However, the equations can be easily extended to include this account, and nothing
substantial would change.
Chapter 6 Saving, Investment, and the Current Account 173

repay their loans. We have so far assumed that defaults do not happen, but
we shall return to that issue at the end of the chapter.

6-5 LIMITATIONSON FOREIGNBORROWING


AND LENDING
So far, when we have discussed the open economy, we have assumed that
residents in one country can borrow from or lend to foreigners on a world
capital market at a given interest rate r. This, of course, is a highly idealized
view. We must now add in three major limitations to our basic framework:
( I ) administrative controls, which limit the access of domestic residents to
foreign capital markets; (2) the effects of the country's own saving and
investment decisions on the world interest rate; and (3) the risk and enforce-
ment problems in foreign borrowing and lending, which limit the extent of
international capital flows.

Administrative Controls
Many governments, especially those in developing countries, impose re-
strictions on the ability of domestic residents to borrow and lend abroad.
Here we look at the basic consequences of these controls and some of the
reasons that such controls are instituted. In later chapters we shall examine
their effects in further detail.
With complete capital controls there could be no borrowing from or
lending t o the rest of the world. The country would live in financial isolation.
Its current account would have to balance in every single period. Domestic
interest rates would bear no relationship to world rates. They would simply
adjust to equilibrate saving and investment as they did in the model of the
closed economy described at the beginning of the chapter.
Let us return for a moment t o Figure 6-4. Without capital controls, the
current account is in surplus at the rate r h , If the government decides to
impose controls, excess domestic saving cannot be used to buy foreign
bonds o r to invest abroad. With saving higher than investment, rh cannot be
the equilibrium interest rate at home. Because the current account has to be
balanced, the domestic rate will have to fall until saving equals investment.
This occurs at the rate r,. For a country that would have a current account
surplus with free capital mobility, the net effect of controls is to reduce
domestic interest rates, raise investment, and lower saving.
By forcing the economy into financial autarky (that is, isolation from
the rest of the world), capital controls can have adverse effects on the level
of economic well-being. We can use the two-period model to illustrate this
quite simply. In Figure 6-9, let E be the endowment point, with the utility
level ULo. If world interest rates are at level r , then the country would like to
borrow in the first period and consume at the point A . This would allow
economic agents to reach the utility level U L , . Instead, the economy must
stay at E because of the capital controls. The same loss in welfare as the
result of capital controls is readily found in the case in which the country
would be in current account surplus in the first period in the absence of
capital controls.
With capital controls in place, the kinds of shocks that we considered
earlier will, in general, affect the domestic interest rate rather than the
174 Part I1 Intertemporal Economics

Period I

Figure 6-9
Capital Controls and the
Economic Well-being of the
Country

current account. For example, a temporary decline in output following a


drought caused a current account deficit in Figure 6-6. Now, the effect is to
raise interest rates, as Figure 6-10 shows.
To summarize, then, the shocks which shift the saving curve to the left
tend to increase the domestic interest rate rather than to worsen the current
account. The same applies for shocks that increase investment possibilities
at home. With full capital controls, a rise in the world interest rate does not
have a direct effect on the domestic interest rate, saving, or investment. By
virtue of its restrictions on capital, the country becomes insulated from
foreign interest-rate shocks.
One crucial policy implication of capital controls involves national
saving policies. Many governments adopt policies to encourage saving (tax
incentives, for example), with the aim of increasing investment. When capi-
tal markets are open, a policy that raises national saving tends to increase
the current account surplus but not domestic investment. In this case, capi-
tal controls might be useful to translate a rise in domestic saving into a rise in
domestic investment.

Large-Country Effects on World Interest Rates


The notion that domestic residents can borrow or lend freely at a given rate r
is based on the assumption that their particular economy is a small part of
the world capital market. This is a good approximation for most countries in

Figure 6-10
A Temporary Output Drop
Under Capital Controls
Chapter 6 Saving, Investment, and the Current Account 175

the world, other than a handful of the largest industrialized economies. Even
a major industrial economy such as the Netherlands represents only 1.6
percent of the total output of the industrialized countries.17 Therefore, even
fairly sizable shifts in this country's saving or capital formation would not
have much effect on world capital market equilibrium. In contrast, the
United States contributes about 36 percent of the total output of the industri-
alized countries. Movements of desired saving and investment in the United
States tend to have significant effects on world interest rates. The same is
true of Japan and Germany, and to a lesser extent in the United Kingdom,
France, Italy, and Canada.I8
The key to understanding large-country effects is to examine the deter-
mination of the world interest rate ( r , ) . In an integrated global capital mar-
ket, r, is determined so that total world saving S, (equal to the sum of saving
in country 1, country 2, and so on, so that S,,, = S I + S2 + . . .) is equal to
total investment (I,, = II + I2+ . . The world as a whole is a closed
0 ) .

economy. Therefore it must be the case that S, = I , .


Let us consider now the case of an economy, say, the United States,
which is large relative to the overall world market. (Following our usual
convention, an unstarred variable refers to the home country, while a starred
variable refers to the rest of the world.) The global equilibrium occurs where

Condition (6.25) states that world investment equals world saving. By


rearranging its terms, we see that this expression is equivalent to saying that
the U.S. current account balance must equal the opposite of the current
account balance of the rest of the world:I9

Figure 6-11 shows the equilibrium world interest rate as that rate at
which the U.S. current account deficit is equal in value to the foreign current
account surplus. If the two regions start in financial autarky, either because
the United States or the rest of the world has capital controls, the equilib-
rium interest rates would be set separately in the two markets. As drawn in
the graph, the domestic rate in the United States ( r , ) would be higher than
the rate in the rest of the world (rz). This is because the United States has
been drawn (realistically!) as having a low saving rate.

l7 Figure for the year 1988, from The World Bank, World Development Report 1990
(New York: Oxford University Press, 1990).
' V h e s e largest seven industrial countries are often called the Group of Seven, or G-7 for
short.
l9 In theory, the sum of the current accounts of all the countries in the world should sum
to zero. In practice, this is not the case. There is, in fact, a "world current account discrep-
ancy," in which the total of the current accounts of all the countries in the world have summed
to a large negative number in recent years, on the order of -$67 billion in 1989 (International
Monetary Fund, Inrernarional Financial Statistics, 1989 Yearbook). This discrepancy is attrib-
uted to a variety of measurement problems, including unrecorded capital flows and under- and
overinvoicing of exports and imports, often for the purposes of smuggling.
176 Part I1 Intertemporal Economics

I S,I I S*, I*

Figure 6-11
Global World Equilibrium of Saving and Investment

If complete capital mobility between the two regions were established,


for example, via a capital market liberalization in the restricted region, a
single world interest rate would result. The domestic interest rate in the U.S.
economy would fall and the rate in the rest of the world would rise until both
rates became the same. Investment would increase and saving would fall in
the United States, and its current account would move into deficit. In the
rest of the world, saving would rise and investment would fall: its current
account in the rest of the world would move into surplus. In the final equilib-
rium, total world saving would equal total world investment, and the U.S.
current account deficit would be exactly matched by the surplus in the rest of
the world.
These two diagrams help us to discover another important point: for a
large country, shifts in saving and investment provoke effects on world (and
domestic) interest rates as well as in the current account. Consider, for
example, a fall in the U.S. saving rate, a s shown in Figure 6-12. (Such a
decline in saving might arise because of a rise in expected future income in
the United States). At the initial interest rate (ro),the decline in saving leads
to an excess of world investment over world saving. World interest rates
therefore rise to r l , where ( I - S) again equals (S* - I*).

Figure 6-12
Global Effects of a Decline in United States Saving

I
s, I I S*, I*
Chapter 6 Saving, Investment, and the Current Account 177

Cases
Free Capital Free Capital
Kind of Mobility Capital Mobility
Shock (small country) Controls (large country)
Rise in the Rise in CA; No effect on CA; Rise in CA;
S curve no effect on r fall in r fall in r .
Rise in the Fall in CA; No effect on CA; Fall in CA;
I curve no effect on r rise in r rise in r
Rise in Fall in CA; No effect on CA; Fall in CA;
(S* - I*) fall in r no effect on r fall in r

The final effect is an increase in the world interest rate and a worsening
of the current account in the United States (from AB to CD), coupled with an
improvement in the current account in the rest of the world (from A'B' to
C'D'). The larger the United States is in world markets, the more adjustment
will occur through a rise in the interest rate. The smaller the United States is,
the more adjustment will come through a deterioration in the U.S. current
account. Thus, the large-country case falls somewhere between the small-
country model and the capital controls case in the effect of the shift in saving
on the current account and on the interest rate.
Table 6-7 summarizes the various cases we have considered here. Each
column in Table 6-7 corresponds to one of the three cases analyzed: a small
country with free capital mobility, a small country with capital controls, and
a large country with free capital mobility. Each row corresponds to a differ-
ent type of shock: an increase in the desired saving in the home country, an
increase in desired investment, and a rise in saving in the rest of the world.
The rest of the table describes the effects for each combination.

Risk and Enforcement Problems


To simplify the analysis, we have assumed so far that all loans are repaid (or
serviced in full in present-value terms). There are at least two reasons why,
in reality, this might not be the case. First, the borrower might become
insolvent, that is, unable to service the debts in full out of the stream of
current and future income. Second, the borrower might choose not to repay
the loans, believing that the costs of nonpayment are less than the burden of
repayment.
Voluntary nonpayment can occur because international loans present a
serious problem of enforcement. It is hard for lenders to collect their loans
when a foreign debtor has a repayment problem, because the problems of
legal enforcement of contracts is particularly difficult when the creditor and
debtor are in different countries. This is especially true of loans to foreign
governments, which are often called sovereign loans, since it is difficult to
178 Part I1 Intertemporal Economics

compel a foreign government to honor a debt. In this case, lenders will not
provide all the funds that the foreign borrower wants at the prevailing inter-
est rate. Rather, they will lend only as much as rhey think can be collected.
When a borrower government has a large external debt, it must grapple
with the choice of repaying the loan versus suspending its debt-service pay-
ments. The government must calculate the benefits of suspending payments
(the foreign exchange that it saves) versus the costs of such action. These
include various penalties for nonpayment, plus the costs of a bad reputation,
which can harm the country in its future dealings with foreign creditors. The
direct penalties that can be imposed by disgruntled creditors include (1) a
suspension of further lending, (2) a withdrawal of short-term lending to
support exports and imports, (3) an attempt t e disrupt the international trade
of the country, and (4) an attempt to disrupt the foreign relations of the
country. These penalties can impose burdens on defaulting countries, but
they do not generally yield much in the way of direct financial benefits for the
lenders.
These penalties help to define the limits of safe lending. If the penalties
for nonpayment are very high, and are known t o be high, then the debtor
government attempts to repay as much as possible, lest the penalties be
incurred. In this case, it is safe t o lend t o a foreign government, since it will
make a strong attempt to repay its loans. If the penalties are small, foreign
governments will not make much of an effort at repayment, so it is rather
unsafe to lend even small amounts.
What is important for us here is that as long as enforcement problems
exist, there will probably be a smaller flow of international lending than there
would be if contracts were perfectly enforceable. At first, residents of a
borrower country will find that they face a higher rate of interest the more
their country borrows from the rest of world, the higher interest rate repre-
senting a risk premium to compensate the lenders for the growing risk of
default. After a certain amount of debt has been incurred, the risks of lending
to the country cannot be compensated by a higher risk premium, and the
country is simply cut off from additional credits.
The full implications of this kind of credit rationing require a thorough
and separate analysis. But in essence, the current account behaves some-
what like the case of a large country: shifts in saving and investment affect
both the current account and the interest rate. (These points are discussed in
more detail in later chapters.)

In an economy with free capital mobility, national saving does not have to
equal national investment. The excess of saving over investment is the cur-
rent account of the balance of payments. The current account balance tends
to be an increasing function of the interest rate because a higher interest rate
tends to increase saving (though the effect is theoretically ambiguous) and to
reduce investment.
A current account surplus also means that a country is accumulating
net international assets; that is, its net claims on the rest of the world are
increasing. A current account dejicit means that a country is decumulating
net international assets. Thus, the current account is also defined as the
change in the net international investment position (NIIP) of a country.
Chapter 6 Saving, Investment, and the Current Account 179

When the NIIP is positive, the country is a net creditor of the rest of the
world, and when it is negative the country is a net debtor. There are two
additional ways to define the current account: first, as the difference be-
tween national income and absorption; and second, as the sum of the trade
account and the service account of the balance of payments.
During the 1980s, the United States was transformed from the world's
largest international creditor into the world's largest debtor as a result of
large and sustained current account deficits. (Nonetheless, data problems
prevent us from getting an exact measure of the net debt position.) Over the
same period, Japan and West Germany ran vast current account surpluses
and became the major international creditors.
Many factors influence the current account. A rise in world interest
rates will tend to improve the CA balance of a small country by raising -
saving and reducing investment. Increased investment prospects (say, be-
cause of a natural resource discovery) tend to reduce the CA balance. A
transitory fall in national income (say, because of a fall in the terms of trade
or an unfavorable harvest), tends to lower the CA balance by reducing
national saving. A permanent decline in national income, however, should
have little or no effect on the current account, since consumption spending
should fall by approximately the same amount as the decline in income. (Of
course, if the permanent shock is widely but wrongly interpreted to be
temporary, then the current account would nonetheless decline.) In general,
the optimal response to supply shocks (either in output levels or the terms of
trade) can be summarized in the phrase "finance a transitory shock; adjust to
a permanent shock. "*O
Countries, like individuals, are bound by an intertemporal budget con-
straint: the discounted value of aggregate consumption must be equal to the
discounted value of national production minus the discounted value of in-
vestment, plus the initial net international investment position. This can be
put another way. If a country is a net debtor, then the economy must run
trade surpluses in the future with a present discounted value equal to the
initial net debt.
Several limitations must be added in to the basic model of borrowing
and lending. First, some governments establish administrative restrictions
(capital controls) to international borrowing or lending. With complete capi-
tal controls, there is no borrowing from or lending to the rest of the world,
and the country must live in financial isolation. Domestic interest rates
would differ from world rates and the current account would have to be zero
every period. Domestic saving would always have to equal domestic invest-
ment.
Second, the basic model of borrowing and lending assumes that the
country is sufficiently small that shifts in its domestic investment and saving
do not affect the world interest rate. This assumption pretty well describes
the case for most countries in the world except for a handful of industrialized
economies. For these large economies, changes in domestic saving and in-
vestment will tend to have significant effects on world interest rates. In an

20 "Finance," here, means to run a current account deficit; "adjust" means to lower
consumption by enough to absorb the shock without borrowing.
180 Part I1 Intertemporal Economics

integrated world capital market, the international interest rate is determined


so that total world saving is equal to total investment.
Third, the basic model assumes that all loans are repaid (or at least
serviced in full in present-value terms). However, some borrowers may
become insolvent (unable to service their debts in full out of current and
future income), while others who could pay may choose to default, knowing
that it is hard for the creditor to force a repayment of the loans. The difficulty
of enforcing loan payments is especially great for sovereign loans, that is,
loans to foreign governments. When potential lenders understand that the
borrower may have an incentive to default in the future, they will restrict the
supply of loans to that borrower to the level that the lender believes will be
repaid.

Key Concepts
balance of payments capital account
current account capital outflow
net international investment capital inflow
position official foreign-exchange reserves
net creditor country errors and omissions
net debtor country national intertemporal budget
absorption constraint
small-country assumption net resource transfer
trade balance capital controls
service balance capital mobility
official development assistance large-country effects
terms of trade sovereign loans

Problems and Questions


1. Country A is a small open economy. Would it be possible for this
country to have an interest rate different from that of the rest of the world?
Why?
2. Countries that run current account surpluses are likely to decrease their
consumption in the future. True or false? Explain.
3. Discuss why the United States shifted during the 1980s from being a
major creditor into being the world's largest debtor.
4. Discuss the relation between an increase in the net holdings of interna-
tional assets, a surplus in the current account, and a positive trade balance.
5. Assume that country B is a net creditor. The value of national saving is
fixed at a certain level, and initially, its current account is zero. What would
happen to the following variables if the value of this country's international
assets goes up because of changes in their valuation?
a. Net international investment position.
b. Current account.
c. Investment.
6. How would Figures 6-4a and 6-4b change if the income effect for savers
becomes larger than the substitution effect above a certain level of the inter-
national interest rate?
7. Describe the effects on the interest rate, domestic savings, and domestic
investment of the following events (analyze the cases of a closed economy, a
Chapter 6 Saving, Investment, and the Current Account 181

small open economy, a large open economy, and an economy with capital
controls.
a. Country C discovers large new reserves of oil. The reserves are highly
profitable, but it will take five years of new physical investments to
bring them into operation.
b. Cold weather in country D forces extensive factory closings for three
months. The lost production cannot be recouped, but production re-
turns to normal by the spring.
c . New synthetic fibers reduce the demand for copper, permanently low-
ering its price relative to other goods. Consider the effect on country E,
a copper exporter.
8. Assume that investment and saving are determined by the following
equations: I = 50 - r; and S = 4r.
a. If the economy is closed, what are the equilibrium levels of the interest
rate, savings, investment, and the current account?
b. How would your answer to (a) change if the country is a small open
economy and the international interest rate is 8 percent? What would
happen if the interest rate rises to 12 percent?
c. How would your answers to (a) and (b) change if the investment func-
tion becomes I = 70 - r?
9. Consider an economy with the following characteristics: production in
period 1 ( e l ) is 100; production in period 2 (Q2) is 150, consumption in
period 1 ( C , )is 120, and the world interest rate is 10 percent. (Assume that
there are no investment opportunities.) In the framework of the two-period
model, calculate
a. The value of consumption in the second period.
b. The trade balance in both periods.
c . The current account in both periods.
10. If there is no final period in which debts have to be repaid, debtor
countries are not constrained by the intertemporal budget constraint. True
o r false? Explain.
11. Assume that Figure 6-1 1 represents the situation of two large countries
in the first period of the two-period model. What would the diagrams look
like in the second period? Which curves would have to shift so that the two
countries maximize their welfare subject to their intertemporal budget con-
straint?
12. How would the following transactions be recorded in the balance of
payments?
a. A U.S. corporation exports $50 million of merchandise, using the pro-
ceeds to open up a foreign factory.
b. U.S. residents receive dividends on holdings of Toyota stocks in the
amount of $10 million.
c . Sylvester Stallone receives $20 million in royalties from overseas box
office sales of Rambo.
d . U.S. holdings in Libya are nationalized without compensation.
182 Part I1 Intertemporal Economics

APPENDIX: BALANCE-OF-PAYMENTS ACCOUNTING


In this section we will study how the balance-of-payments accounts of a
country are actually measured. The current account is measured over a
specified interval of time, usually a month, a quarter, or a year. For a typical
country during one of these time intervals, there are millions of transactions
of individual households, firms, and the government, which must be summed
up to calculate the overall current account balance.
The basic idea of balance-of-payments accounting relies on the fact
that there are two definitions of the current account: as the trade balance
plus net factor payments from abroad and as the change in the country's net
foreign investment position. Imbalances in trade have as their counterpart an
accumulation or decumulation of net international assets. The basic method
of balance-of-payments accounting takes advantage of the fact that trade
flows and financial flows are two sides of each transaction.
In the balance-of-payments accounts, the transactions are divided be-
tween current flows (exports, imports, interest receipts, etc.) and capital
flows (changes in ownership of financial assets), as shown in Table A-1. The
top part of the table is sometimes called simply the current account items,
while the bottom part of the table measures the so-called capital account. In
principle, the current account and the capital account must be identical in
value, when the change in international reserves is included in the capital
account. In practice, because of errors and omissions in the actual recording
of transactions, the items in the current account do not always sum to the
same amount as the items in the capital account.

1. Current Account ( C A = 1.1 + 1.2 + 1.3)


1.1 Trade balance
Exports of goods
Imports of goods
1.2 Service balance
Nonfactor services (freight, insurance, tourism, etc.)
Capital services (interest receipts, profit remittances)
Labor services (worker's remittances)
1.3 Unilateral transfers
2. Capital Account (CAP = 2.1 + 2.2 + 2.3)
2.1 Net foreign investment received
2.2 Net foreign credits received
Short term
Long term
3. Errors and Omissions
4. Balance-of-payments result (BP = 1 + 2)
(= change in net official international reserves)
Chapter 6 Saving, Investment, and the Current Account 183

In theory, all transactions that affect the current account require two
entries in the table. For example, consider the accounting in the U.S. bal-
ance of payments when a West German firm sells $10 million of machinery to
a U.S. importer, which the importer pays for by writing a $10 million check
to the West German firm, which the firm deposits in its U.S. bank account.
There are two parts to the transaction: the shipment of goods, recorded
under "imports," and the payment of the check, which increases the U.S.
bank balances of the West German firm, and thus is recorded under "change
in liabilities to foreigners."
The accounting conventions for these two items are determined so that
the pair of transactions sums to zero. In particular, imports are given a
negative sign in the table (that is, the transaction would be entered as -$I0
million); the increase in liabilities to foreigners is recorded as a positive entry.
in the capital account (the transaction would be entered as + $ l o million). If
this were the only transaction to be considered, the current account deficit
would equal -$lo million, while the capital account balance (the mirror
image of the current account balance) would equal + $ l o million.
An increase in the country's net foreign assets B*, which can mean a
rise in claims against foreigners or a fall in liabilities owed to foreigners, is
called a capital outflow. A decrease in net foreign assets is called a capital
inflow. Thus, in the transaction we have been describing, we can say that
there is a capital inflow that is financing the current account deficit in the
United States. Alternatively, we can say that a surplus in the capital account
is financing a deficit in the current account.
The following conventions apply to balance-of-payments accounting:
1. Export earnings and receipts of interest from abroad are entered as
positive items in the current account.
2. Import payments and payments of interest on foreign liabilities are
entered as negative items in the current account.
3. Increases in claims against foreigners and decreases in liabilities to
foreigners (capital outflows) are entered as negative items in the
capital account.
4. Decreases in claims against foreigners and increases in liabilities to
foreigners (capital inflows) are entered as positive items in the capi-
tal account.
If all transactions in the balance of payments were actually recorded as
they occur, according to the conventions just outlined, the balance of pay-
ments would sum to zero (adding up the current account and the capital
account). As we will note, however, some transactions are recorded only in
part, so that the statisticians that prepare the balance of payments are some-
times left without a complete record of transactions, with the result that the
recorded items of the current and capital accounts do not add up to exactly
zero.
Consider the following set of transactions. Study carefully how they
are recorded in the accompanying Table A-2. (Each transaction is lettered,
and the entries in the balance of payments are recorded with the transaction
number shown in parentheses.)
(a) A U.S. exporter ships $5 million of grain to the Soviet Union on a
90-day credit (in other words, the Soviet importer owes the $5 mil-
lion in 90 days).
184 Part I1 Intertemporal Economics

Current Account
Trade balance
Exports +5
Imports
Services
Interests
Dividends
Other
Unilateral transfers
Capital Account
Net foreign investment
Net credits
Short term -5
Long term
Balance of Payments

(b) A U .S. individual receives a dividend payment of $1 million from


a factory that he owns abroad, and he uses the money to reinvest in
the factory.
(c) Following an earthquake in Armenia, U.S. private relief agencies
send $15 million worth of first aid equipment and clothing.
(d) A Japanese firm imports $20 million of oil from Saudi Arabia,
paying for the oil by writing a check on its account at the New York
branch of Chase Manhattan Bank. The check is deposited in the
Saudi Arabian account at the same bank. (In this case, no entry is
made, as the transaction does not affect the U.S. balance of pay-
ments .)
(e) A U.S. importer buys $10 million of merchandise from a Japa-
nese electronics firm and pays for the transaction with a loan from a
Japanese bank that finances the deal.
(f) The U.S. Treasury sells official reserve holdings of Deutsche
marks to U.S. securities houses for $20 million in cash.
Note that for each capital transaction, care must be taken to record the
transaction in the correct subcategory in the capital account. Distinctions
are made between short-term and long-term capital, with bank balances, for
example, constituting a form of short-term capital and long-term bonds and
Chapter 6 Saving, Investment, and the Current Account 185

equities constituting a form of long-term capital.' Long-term capital is subdi-


vided further into securities and foreign direct investment, with the latter
signifying direct ownership and control over a firm operating in a foreign
country (or foreign ownership and control over a firm operating in the
United States).
Another crucial distinction is made between financial assets owned (or
owed) by the government versus those owned (or owed) by the private
sector. The central banks of most countries (and sometimes the treasuries as
well) maintain holdings of short-term foreign assets, such as short-term
Treasury bills issued by foreign governments. These holdings are called the
officialforeign-exchange reserves of the central bank. We shall see in later
chapters how these reserves may be used by the central bank to help manage
the exchange rate of the domestic currency, through a process in which the
central bank buys and sells foreign reserves in return for domestic currency
held by the public.
Because of the importance of foreign-exchange holdings for the ability
of a government to manage the exchange rate, special care is given to ac-
counting for changes in foreign-exchange reserves. The ofJicial reserve
transactions balance measures the change in the net official foreign reserve
balances of the government. The country is said to have a positive balance if
the government is accumulating net international reserves and a negative
balance if official reserves decline within the period. Note how this concept
is related to the current account and the capital account. If we measure in
the capital account all capital items except the official holdings of foreign
exchange, then we have the following:
Official reserve transactions balance = change in net official foreign
reserves
= current account + nonofficial
capital account
By adding up the current account and all capital account items except the
official reserves, we get the official reserve balance, with the sign convention
that a positive value indicates an increase in net foreign reserves.
The official reserve balance is sometimes loosely called the "overall
balance of payments." Countries are said to have an "overall" surplus if
they are accumulating official reserves and an overall deficit if they are
loosing reserves.
Note that from an economic point of view, it might be more accurate to
say that the overall balance of payments is always zero. When we count
changes in official reserves as part of the capital account and use the earlier
convention that increases in net claims against the rest of the world are
entered into the accounts with a negative sign (so that reserve increases are
measured negatively), then it remains true that the current account and the
capital account will always sum to zero. The "overall" balance can be

In the U.S. balance of payments, long-term assets and liabilities are financial claims
with an original maturity of 1 year or more. Thus, for example, a 20-year asset issued 191h years
ago and coming due in half a year is considered a long-term asset for purposes of the balance-of-
payments accounts.
186 Part I1 Intertemporal Economics

different from zero only when we separate the changes in official reserve
balances from the capital account; then, the current and nonreserve capital
account do not have to sum to zero. This is why we create a category
"change in net official international reserves," item 3 in Table A-1.
There is one additional reason why the overall balance of payments
might not sum to zero: statistical discrepancies. If each international trans-
action were separately recorded, the accounts would necessarily balance for
the reasons shown. But in fact, the transactions are not recorded one-by-one
as they occur. Statisticians that prepare the international accounts often
observe the trade flows (exports and imports) and the capital flows (changes
in net claims against the rest of the world) separately. They receive reports
from customs about the international flows of goods, and separate reports
from the financial markets about the changes in claims and liabilities vis-a-
vis foreigners. But many transactions, both in trade and in financial assets go
unreported, both because of the huge number and complexity of the transac-
tions involved, and because of deliberate attempts to avoid detection (as in
the case of money laundering or tax evasion).
For this reason, a separate item must appear on the balance-of-pay-
ments accounts, called "Errors and omissions," or "balancing item." It is
given the value equal to the opposite of the sum of the current account plus
the capital account (including official reserves), so that in fact the sum of all
items in the accounts-including errors and omissions-equals zero. In the
United States, the errors and omissions account was large and positive for
many years in the 1980s. This suggested to many observers that foreigners
were accumulating assets in the United States, but not reporting the increase
in such assets to the authorities. Thus, the measured capital inflows were too
small, and what would have been a larger positive entry in the capital ac-
count (measuring the accumulation of foreign claims on U.S. residents)
became a positive entry in the errors and omissions line of the balance of
payments.
Countries face quite diverse circumstances regarding capital account
flows and their net foreign asset positions. For comparative purposes, we
can briefly analyze the cases of Japan and the United States during the
1980s. As we have already seen, the United States shifted into substantial
deficits in the current account, while Japan experienced surpluses during the
period. As a counterpart to these current account developments, there has
been a capital account surplus in the United States and a capital account
deficit in Japan. In other words, foreigners have been accumulating claims of
different forms on the United States (Treasury bonds, stocks, real estate,
productive companies), which has implied a capital inflow (capital account
surplus). By contrast, Japan has been a net investor in the rest of the world:
its purchase of foreign assets represent a capital outflow (capital account
deficit). Table A-3 shows the capital account (a flow) and the net foreign
asset position (a stock) for the two countries2

We should stress once again that while the data in this table show an unmistakable
trend, in which the U.S. net foreign investment position is falling and that of Japan is rising, the
specific magnitudes in the table are subject to several measurement problems.
186 Part I1 Intertemporal Economics

different from zero only when we separate the changes in official reserve
balances from the capital account; then, the current and nonreserve capital
account do not have to sum to zero. This is why we create a category
"change in net official international reserves," item 3 in Table A-1.
There is one additional reason why the overall balance of payments
might not sum to zero: statistical discrepancies. If each international trans-
action were separately recorded, the accounts would necessarily balance for
the reasons shown. But in fact, the transactions are not recorded one-by-one
as they occur. Statisticians that prepare the international accounts often
observe the trade flows (exports and imports) and the capital flows (changes
in net claims against the rest of the world) separately. They receive reports
from customs about the international flows of goods, and separate reports
from the financial markets about the changes in claims and liabilities vis-a-
vis foreigners. But many transactions, both in trade and in financial assets go
unreported, both because of the huge number and complexity of the transac-
tions involved, and because of deliberate attempts to avoid detection (as in
the case of money laundering or tax evasion).
For this reason, a separate item must appear on the balance-of-pay-
ments accounts, called "Errors and omissions," or "balancing item." It is
given the value equal to the opposite of the sum of the current account plus
the capital account (including official reserves), so that in fact the sum of all
items in the accounts-including errors and omissions-equals zero. In the
United States, the errors and omissions account was large and positive for
many years in the 1980s. This suggested to many observers that foreigners
were accumulating assets in the United States, but not reporting the increase
in such assets to the authorities. Thus, the measured capital inflows were too
small, and what would have been a larger positive entry in the capital ac-
count (measuring the accumulation of foreign claims on U.S. residents)
became a positive entry in the errors and omissions line of the balance of
payments.
Countries face quite diverse circumstances regarding capital account
flows and their net foreign asset positions. For comparative purposes, we
can briefly analyze the cases of Japan and the United States during the
1980s. As we have already seen, the United States shifted into substantial
deficits in the current account, while Japan experienced surpluses during the
period. As a counterpart to these current account developments, there has
been a capital account surplus in the United States and a capital account
deficit in Japan. In other words, foreigners have been accumulating claims of
different forms on the United States (Treasury bonds, stocks, real estate,
productive companies), which has implied a capital inflow (capital account
surplus). By contrast, Japan has been a net investor in the rest of the world:
its purchase of foreign assets represent a capital outflow (capital account
deficit). Table A-3 shows the capital account (a flow) and the net foreign
asset position (a stock) for the two c ~ u n t r i e s . ~

We should stress once again that while the data in this table show an unmistakable
trend, in which the U . S . net foreign investment position is falling and that of Japan is rising, the
specific magnitudes in the table are subject to several measurement problems.
Chapter 6 Saving, Investment, and the Current Account 187

Japan United States


Net Foreign Net Foreign
Assets Capital Assets Capital
Year Position Account Position Account

Source: For Japan, Management and Coordination Agency, Japan Statistical Yearbook,
various issues; for the United States, Economic Report of the President, 1990, and Depart-
ment of Commerce, Survey of Current Business (Washington, D . C . : U . S . Government
Printing Office,June 1990).

Appendix Summary
The balance-of-payments records all (known) transactions between a coun-
try's residents and the rest of the world. Balance-of-payments accounting
relies on two different definitions for the current account: the trade balance
plus net factor payments from abroad and the change in the country's net
international investment position. Trade flows and financial flows are two
sides of each transaction. Thus, trade imbalances have as their counterpart
an accumulation or decumulation of net international assets. An increase in
the country's net foreign assets is called a capital outflow; a decrease in net
foreign assets is called a capital inflow. In principle, double-entry bookkeep-
ing ensures that a current account surplus (deficit) is matched by an identical
capital account deficit (surplus). In practice, the current account and capital
account may differ because of errors and omissions in data collection.